[saurik]

The LGPL clearly states that a Combined Work which includes the the Library must "1) Use a suitable shared library mechanism for linking with the Library. A suitable mechanism is one that (a) uses at run time a copy of the Library already present on the user's computer system, and (b) will operate properly with a modified version of the Library that is interface-compatible with the Linked Version." as well as insisting that the terms of the license under which you distribute the Application "effectively do not restrict modification of the portions of the Library contained in the Combined Work and reverse engineering for debugging such modifications" <- taken together, maybe (maybe!) you could make an argument that your shared library loader was legit while the code using that shared library loader was evil (though that clearly violates the intention of this license in a way that is so blatant I would be shocked if a judge or a jury didn't shake their heads at your claim), but then the rest of the anti-virus software wouldn't be able to be distributed under a typical commercial license as modifications and reverse engineering of that code would have to be allowed.

[AstralStorm]

I always wondered whether code signed deployment is not compliant with LGPL and where the threshold lies.

Specifically, if forcing to sign a package with a different key (making it a different package) for private purposes is enough, or if the redistribution rights of the whole is required. Finally, if you cannot replace the software because of code signing and no public debug mode, that seems incompatible too...

[kevin_b_er]

You are quoting LGPLv3, which has provisions for making so you can't restrict replacing the library.

7-zip is licensed under LGPLv2.1. "do not restrict" is not a string that appears in v2.1 The entire second part you quoted was added in version 3.